Age, growth and mortality of southern rays bream Brama australis

Revista de Biología Marina y Oceanografía
Vol. 48, Nº3: 585-590, diciembre 2013
DOI 10.4067/S0718-19572013000300014
Article
Age, growth and mortality of southern rays bream
Brama australis (Bramidae) off the
southeastern Pacific coast
Edad, crecimiento y mortalidad natural de la reineta Brama australis (Bramidae)
en la costa este del Pacífico sur
Ciro Oyarzún1, Nelson Cortés2 and Elson Leal3
1
Departamento de Oceanografía, Sección Pesquerías, Universidad de Concepción, Concepción, P.O. Box 160-C, Chile
Unidad de Recursos Hídricos, Servicio de Evaluación ambiental, Ministerio de Medio Ambiente, Lincoyan N°145,
Concepción, Chile
3
División de Investigación Pesquera, Instituto de Fomento Pesquero, Blanco 839, Valparaíso, Chile. [email protected]
2
Resumen.- Se determinó la edad, función de crecimiento y mortalidad natural de la reineta Brama australis provenientes
de capturas comerciales realizadas en la zona central de la costa de Chile. Los resultados indican que la especie crece
rápidamente durante el primer año y alcanza la máxima longitud asintótica (59 cm) aproximadamente al octavo o noveno
año de edad. La función de crecimiento de von Bertalanffy describió satisfactoriamente el crecimiento en longitud a
través de la edad de B. australis, determinada mediante el análisis de los otolitos. Los parámetros estimados para ambos
sexos combinados fueron: en longitud, L∞ = 58,95 cm, K = 0,277 año-1, to = -0,371 años; y en peso, W∞ = 2508,8 g, K = 0,257
año-1, to = -0,712 años. No se encontraron diferencias en la longitud a la edad entre sexos. La tasa de mortalidad total (Z)
desde el análisis de la edad máxima y curva de captura fue relativamente alta (0,9 año-1). La tasa de mortalidad natural
(0,45 año-1) es consistente con los parámetros de vida de un pez pelágico de tamaño medio, de explotación relativamente
baja en el periodo en que fueron colectadas las muestras.
Palabras clave: Bramidae, edad y crecimiento, mortalidad, Pacifico sureste, reineta
Abstract.- The age and growth function and mortality were determined for southern rays bream Brama australis caught in
commercial trawls off the coast of central Chile. The results show that this species grows quickly in the first year and
reaches the maximum asymptotic length (59 cm) at approximately 8 or 9 years old. The von Bertalanffy growth function
described B. australis growth satisfactorily with length-at-age determined using otolith analysis. Parameters for the entire
population (males and females) were: for length, L∞ = 58.95 cm, K = 0.277 yr-1, to = -0.371 yr; and for weight, W∞ = 2508.8 g, K
= 0.257 yr-1, to = -0.712 yr. No differences were found in length-at-age between sexes. Estimates of total mortality rates (Z)
from maximum age and catch-curve analyses were relatively high (0.9 year-1). The natural mortality rate (M = 0.45 yr-1) is
consistent with the life parameters of middle-sized pelagic fish with relatively low exploitation at the time of sampling.
Key words: Bramidae, age and growth, mortality, eastern South Pacific, southern rays bream
INTRODUCTION
The Southern rays bream, Brama australis Valenciennes,
1838, is a member of the family Bramidae that is exploited
by the artisanal fleet along the Chilean coastline.
According to Mead (1972) and Last & Baron (1994), this
family inhabits epipelagic ecosystems at depths over 200
m, although it has been reported at depths exceeding 400
to 500 m. B. australis is widely distributed in the South
Pacific Ocean; it has an apparent circumpolar distribution
in the hemisphere and is restricted to the circulation
boundaries of the subtropical water masses (35°-48°S)
(Pavlov 1991a).
Until a few years ago, B. australis was not considered
to be a commercially important species. Rather, it was
caught as bycatch during trawling for Patagonian
grenadier (Macruronus magellanicus Lönnberg, 1907)
and Chilean jack mackerel (Trachurus murphyi Nichols,
1920) (Muñoz et al. 1995). However, B. australis is
currently highly valued for human consumption in central
Chile. Official records were begun for B. australis catches
off the Chilean coast in 1994 (SERNAPESCA 2011). The
mean landings have increased from 5000 tons (t) between
1994 and 2000 to over 15000 t since 2010 (Fig. 1). The
Vol. 48, Nº 3, 2013
Revista de Biología Marina y Oceanografía
585
Organisms were aged by counting the number of
growth marks found in the otoliths. Prior to aging the
individuals, a periodicity analysis was done to evaluate
the formation of annual growth bands. The number of
translucent zones on each otolith and the nature of the
border were recorded.
Figure 1. Total landings of southern rays breams registered in Chile
from 1994 / Desembarques totales de reineta registrados en
Chile desde 1994
fishery shows seasonality with higher catches during
spring-summer from October to April.
Information on the biology and ecology of B. australis
is limited despite its newfound ecological and commercial
importance for the artisanal fleet. Pavlov (1991a, b, 1994)
performed morphometric, taxonomic, ecological and
reproductive studies on members of the Brama genus
from southeastern Pacific Ocean. Muñoz et al. (1995) and
Garcia & Chong (2002) studied the feeding habits of B.
australis off the coast of central Chile, characterizing this
species as a secondary consumer that preys mostly on
crabs, squids, and small fishes, as it has been reported in
other species of Brama (Seki & Mundi 1991, Pearcy et al.
1993). Leal & Oyarzún (2003) described B. australis
reproduction along the Chilean coast, reporting higher
reproductive activity in winter and an active female
maturation process over 35 cm fork length.
Given the lack of reliable information on biological
parameters for this species, the objetive of this work was
to study growth and mortality in the eastern South Pacific
population of B. australis based on otolith age analysis
of annual increments in their otoliths.
MATERIALS AND METHODS
Sampling took place between 35°20’S and 38°50’S off the
eastern South Pacific coast (Fig. 2). Specimens were
collected monthly from November 1994 to December 1996
from pelagic commercial fisheries. Fishes were sexed
macroscopically, and measured for total length (LT, cm)
and total weight (WT, g). Sagittal otoliths were taken from
each specimen and stored in dry bags. Prior to analysis,
each left otolith was hydrated in distilled water from 24 to
48 h. The external surface of each otolith was examined
microscopically under reflected light.
586 Oyarzún et al.
Age, growth and mortality of Brama australis
To assess the precision of aging, all otoliths were read
twice by 2 different readers and the agreement between
and within readers was evaluated by percent agreement
methods (Beamish & Fournier 1981). Lack of consensus
resulted in the otolith being rejected as «noninterpretable» and excluded from analysis.
After age determination, parameters of the von
Bertalanffy growth function (BGF, Ricker 1975) in length,
Lt = L∞ (1- exp (–k (t - to)) and weight, Wt = W∞ (1- exp (–k
(t - to))b were obtained. Where Lt and Wt are the length
(cm) and weight (g) of fish aged t (years) respectively, L∞
and W∞ are the theoretically maximum length (cm) and
weight (g) respectively, K is coefficient of growth, t0 is
theoretical age at which the length of fish is 0 and b is
slope parameter in the length-weight relationship.
The parameters of BFG were fitted using non-linear
regression (the NONLIN module of SYSTAT v.7,
Wilkinson 1988) with a Gauss-Newton algorithm, where
growth parameters and the residual sum of squares (RSS)
were obtained from each iteration (1000) through quadratic
minimization. Growth curve parameters were compared
between sexes with RSS analysis modified for non-linear
minimum squares (Chen et al. 1992).
A length-weight relationships (W = a*TLb) was fitted to
the data and utilized to calculate the parameters used in BGF
for weight. TL is a total length, a and b are regression
coefficients. The estimated value a that came from the
logarithmic scale (Smith 1990) was transformed to an
arithmetic function. The growth performance index f =
log10(K) + 2/3 log10(W∞) was computed from the growth
parameters for comparing with another species (Sparre &
Venema 1997).
A catch curve-at-age was constructed by expanding
the length-frequency distribution at abundance caught
at age. The total mortality rate (Z) for both sexes was
determined from the descending limb of the curve
according to Sparre & Venema (1997).
The lack of data concerning fishing effort and the
unavailability of data on unfished areas made it not
possible to make a direct estimation of natural mortality
rate (M). Instead, empirical models based on the
periods in which opaque edges decreased, showing one
peak per 12-month period and thereby demonstrating that
the translucent zones (hialine bands) are formed annually.
Individuals of both sexes showed age range between
2 to 8 years i.e., from 2 to 8 marks (hialine bands). In terms
of numbers, most males were younger (2-4 years) and
most females older (5-7 years). For both sexes, 4 yearsold specimens showed the highest frequency. The VBFG
equations for male, female and for pooled are showed in
the Fig. 3.
Observed lengths did not vary greatly with age. Growth
was fast in the first years of life and slow from year 7
onward. Southern rays bream achieve 50% of their L∞ at
age 2, apparently one year before the recruitment to the
fishery.
No differences were found in growth curve parameters
for length between the sexes (P > 0.05). We did not observe
large numbers of either the youngest or oldest fish.
Figure 2. Geographic location of the study on the central coast of
Chile. The black points indicate the approximate sampling sites /
Localización geográfica del estudio en la costa central de Chile.
Los puntos negros indican los sitios aproximados de muestreo
relationships between M, growth parameters and mean
habitat temperature (T°C) were used to obtain approximate
estimates (Pauly 1980). Additionally, Alagaraja (1984),
proposed an alternative method of estimating M by
relating to the natural life span of fish which was defined
as an age at which 1% of a cohort is still alive (Tl, years).
This approach was also used in order to provide an
estimate of the natural mortality rate in B. australis.
The weight of the southern rays bream varied greatly
with age. As it was found for age-length data, growth
was rapid in the first 4 years. Although it slowed
thereafter, the fish still grew appreciably in weight until
10-11 years, at which point growth substantially slowed.
Hence, the fish grew apparently more slowly in weight
than in length. The BGF equation for pooled data was: Wt
= 2508.8 (1- exp (-0.257 (t+0.712))2.84. Table 1 shows the
estimated parameters, asymptotic standard errors (ASE),
and 95% confidence intervals. The growth performance
of B. australis was φ = 1.67. A pooled length-weight
regression (Fig. 4) with a fitted value for a was developed
(R2 = 0.97; n = 425; P < 0.01). The slope of the regression
line (b = 2.84; SE = 0.067) was significantly different from
3.0 (P < 0.05), indicating allometric growth.
Table 1. Summary of estimated parameters for the von Bertalanffy
growth equation of total length (cm) and total weight (g) of Brama
australis for the pooled data / Resumen de los parámetros
estimados por la ecuación de crecimiento de von Bertalanffy para
la longitud total (cm) y peso total (g) de Brama australis para el
total de los datos
RESULTS
A total of 534 adult breams were analyzed. Individuals
ranged from 21 to 58 cm TL. Of the total specimens
measured, 109 were discarded due to otolith breakage or
loss or because they were classified as non-interpretable,
as defined in the methods section.
The edge analysis showed that the percentage of
translucent (hyaline) edges increased markedly in those
Vol. 48, Nº 3, 2013
Revista de Biología Marina y Oceanografía
587
DISCUSSION
This paper presents new information about important
biological aspects of Brama australis in the eastern South
Pacific and the results may be used to prove potential
temporal changes in growth parameters and natural
mortality.
The reports of biological and ecological data for B.
australis are restricted to taxonomical studies (Pavlov
1991a, b), feeding ecology (Muñoz et al. 1995), and
reproduction dynamics (Pavlov 1994, Leal & Oyarzún
2003). Furthermore, several studies of population
dynamics have targeted congeneric species of Brama
inhabiting other latitudes for purposes of commercial
exploitation and management. This interesting fishery has
been explored off the coasts of Australia, New Zealand,
Spain, and Portugal (Seki & Mundi 1991, Savinykch 1993,
Last & Baron 1994).
The generalized VBGF satisfactorily described the
growth of B. australis, and the similar growth curves for
males and females provided a graphic representation of
the lack of significant differences in mean length-at-age
by sex. A slight difference was observed between sexes
in terms of length, possibly due to different proportions
of length in the sampled individuals: despite a 1:1
proportion of sexes, females tended to be slightly larger
than males.
Figure 3. Length growth curve for females, male and for combined
sexes of southern rays breams / Curva de crecimiento en longitud
para hembras, machos y sexos combinados de reineta
The total mortality rates (Z) were determined from the
descending limb of the resulting curve for catch-at-age,
considering a maximum observed age of 11 years for the
combined sexes. A regression estimate obtained from the
slope of a catch-curve truncated at older ages (Fig. 5)
was Z = 0.9 yr -1. On the other hand, using growth
parameters for pooled data, M was estimated in 0.46 yr-1
following Pauly (1980). The mean habitat temperature of
14°C was assumed for calculating M with this method.
The M value obtained with Alagaraja (1984) method was
0.44 yr-1, using 11 years to Tl. A compromise value of 0.45
yr-1 was assumed to M in B. australis.
588 Oyarzún et al.
Age, growth and mortality of Brama australis
Interestingly, in this study most of the sampled
specimens were adults. The lack of juveniles in the study
could result in overestimated values of the growth
parameters (Francis & Campana 2004). Nevertheless, this
result agrees with the reports about other species of
Brama (Pavlov 1991b, Savinykh 1994,1995; Seki & Bigelow
1996), indicating the possibility of differential bathymetric
distributions for juveniles (pelagic-coastal) and adults
(oceanic-demersal), or the lack of selectivity of the gears
for sampling this etary group. On the other hand, perhaps
the strongest alternative explanation for the lack of youngof-the-year lies in the migratory patterns observed in
many bramids. Specifically, breams undergo long-range
migrations along the oceanic water mass according to
circumtropical circulation (Seki & Mundi 1991). Pavlov
(1994) suggests a feeding migration of B. australis from
north to south in summer-autumn and from south to north
in winter-spring to spawn.
Our results about age (2 to 8 years) in B. australis
observed ranged from 21 to 58 cm TL are consistent with
Pavlov (1994) which suggested one to 6 age group to a
few sample number of 13-48 cm B. australis from southeast
Figure 4. Length-weight relationship curve of southern rays breams
fitted to an exponential model / Curva de la relación longitudpeso de reineta ajustada por un modelo exponencial
Figure 5. Catch-curve and estimate of total mortality for southern
rays breams / Curva de captura y tasa de mortalidad total
estimada para reineta
Pacific Ocean. For Brama japonica, Savinykh & Vlasova
(1994) examined whole stained otoliths of 14-54 cm from
the North Pacific and found a maximum age of 9 years,
with 3 and 4 year fish predominating in the catch. For the
same species, rapid growth is observed during the first 3
years, when they reach 80% L∞ (Pearcy et al. 1993). This
species reaches its asymptotic length at approximately 9
to 11 years of age. Lobo & Erzini (2001) used readings of
whole otoliths for 32-56 cm B. brama from Portugal to
identify age groups 3 to 12 years.
Thus, considering the time in which samples were
obtained, when the resource was still subject to lowmoderate exploitation rates, the results of this report
provide an important point of comparison with current or
future studies. For example, to assess possible fishing
effects on individual growth rates and M. The parameters
may also be used in stock assessment analysis providing
management advice for fishery along the Chilean coast.
ACKNOWLEDGMENTS
Both the growth trends and the age and growth
parameters obtained in this study agree with those
reported for B. japonica and B. brama with a maximum
asymptotic length observed near 60 cm (TL), although
only 4 of the specimens used in this study were larger
than 55 cm LT.
We thank the workers of the Instituto de Fomento
Pesquero (IFOP) for helping us obtain samples. Special
thanks to Mrs. Ana Lara for assistence with otolith
reading and to Dr. Rubén Roa for helpful comments on
the manuscript.
The previous background support maximum ages of 8
to 12 years for species of Brama and seem reasonable for
a middle size of pelagic fish. This also supports the annual
formation of translucent zones (hialine bands) observed
in this study.
LITERATURE CITED
This work is the first study on age, growth, and
mortality for Brama australis in the Chilean coast based
on otolith analysis, so no comparison is possible with
other studies of this species. When compared with the
total mortality rates (Z), estimates of M are consistent.
The yield, explotation rates, and F/M ratio are lower than
those reported for other species exploited in this area
(Böhm et al. 1995). However, the natural mortality rate
seems to be consistent with the life parameters of middlesized pelagic fishes with relatively low exploitation at the
time of sampling.
Alagaraja K. 1984. Simple methods for the estimation of
parameters of assessing exploited fish stocks. Indian Journal
of Fishes 31: 177-208.
Beamish R & D Fournier. 1981. A method for comparing the
precision of a set of age determination. Canadian Journal of
Fisheries and Aquatic Sciences 38: 982-983.
Böhm G, H Muñoz, V Bocic, R Serra, M Nilo, M Donoso,
MA Barbieri & G Rosson. 1995. Diagnóstico de las
principales pesquerías nacionales pelágicas zona centro sur1994, 121 pp. Corporación de Fomento de la Producción.
Instituto de Fomento Pesquero, Valparaíso.
Chen Y, D Jackson & H Harvey. 1992. A comparison of von
Bertalanffy and polynomial functions in modeling fish
growth data. Canadian Journal of Fisheries and Aquatic
Sciences 49: 1228-1235.
Vol. 48, Nº 3, 2013
Revista de Biología Marina y Oceanografía
589
Francis R & S Campana. 2004. Inferring age from otolith
measurements: a review and a new approach. Canadian
Journal of Fisheries and Aquatic Sciences 61: 1269-1284.
Ricker WE. 1975. Computation and interpretation of biological
statistics of fish populations. Journal of the Fisheries
Research Board of Canada 191: 1-382.
Garcia M & J Chong. 2002. Composición de la dieta de Brama
australis Valenciennes 1837 en la zona centro sur de Chile
(VIII región) en otoño de 2000 y verano de 2001. Gayana
66: 225-230.
Savinykh VF. 1993. Some ecological characteristics of young
pomfret in the Northwest Pacific. Marine Biology 2: 7783.
Last P & M Baron. 1994. Rays bream- a new pelagic fishery?
Australian-Fisheries 53: 19-22.
Leal E & C Oyarzún. 2003. Talla de madurez y época de
desove de la reineta (Brama australis Valenciennes, 1836)
en la costa central de Chile. Investigaciones Marinas 31:
17-24.
Lobo C & K Erzini. 2001. Age and growth of Ray’s bream
(Brama brama) from the south of Portugal. Fisheries
Research 51: 343-347.
Mead GW. 1972. Bramidae. Dana-Report 81: 1-175.
Muñoz G, N Cortés, M Arriaza & C Oyarzún. 1995.
Comportamiento trófico de una especie poco conocida,
Brama australis Valenciennes, 1837 (Pisces: Bramidae).
Biología Pesquera 24: 51-55.
Pauly D. 1980. On the inter-relationships between natural
mortality, growth parameters and mean environmental
temperature in 175 fish stocks. ICES 39: 175-192.
Pavlov YP. 1991a. Brama australis Valenciennes-A valid
species of Sea Bream (Bramidae) from Southeastern Pacific
Ocean. Journal of Ichthyology 31: 6-9.
Pavlov YP. 1991b. Information on morphometrics and ecology
of pomfrets of the genus Brama inhabiting the southeastern
Pacific Ocean. Journal of Ichthyology 31: 120-124.
Pavlov YP. 1994. Data on ecology of Sourthern pomfret, Brama
australis in the southeastern Pacific Ocean. Journal of
Ichthyology 34: 124-126.
Savinykh VF. 1994. Size-age structure and growth rate of the
Pacific Pomfret Brama japonica. Voprosy Ikhtiology 34:
494-500.
Savinykh VF. 1995. Migrations of the Pacific Pomfret Brama
japonica. Russian Journal of Marine Biology 20: 271-277.
Savinykh V & L Vlasova. 1994. Size-age structure and growth
rate of the Pacific pomfret Brama japonica (Bramidae).
Journal of Ichthyology 34(8): 97-107.
Seki M & K Bigelow. 1996. Aspects of the life history of the
Pacific Pomfret, Brama japonica, during winter occupation
of the Subtropical Frontal Zone. International Bulletin of
the North Pacific Commission 53: 207-215.
Seki M & B Mundy. 1991. Some notes on the early life stages
of the Pacific Pomfret, Brama japonica, and other Bramidae
from the Central North Pacific Ocean. Japanese Journal of
Ichthyology 38: 63-68.
SERNAPESCA. 2011. Anuario estadístico de pesca, 291 pp.
Ministerio de Economía Fomento y Reconstrucción,
Santiago.
Smith ST. 1990. Use of statistical models for the estimation
of abundance from groundfish trawl survey data. Canadian
Journal of Fisheries and Aquatic Sciences 47: 894-903.
Sparre P & S Venema. 1997. Introduction to tropical fish
stock assessment. FAO Fisheries Technical Report 306: 1337.
Received 30 January 2013 and accepted 18 October 2013
Associate Editor: Mauricio Landaeta D.
590 Oyarzún et al.
Age, growth and mortality of Brama australis